American Journal of Geographic Information System 2017, 6(2): 64-82 DOI: 10.5923/j.ajgis.20170602.03

Spatial Monitoring of Urban Growth Using GIS and Remote Sensing: A Case Study of Metropolitan Area,

Monica Machuma Katyambo*, Moses Murimi Ngigi

Institute of Geomatics, GIS and Remote Sensing, Dedan Kimathi University of Technology, , Kenya

Abstract Nairobi city expanded spatially leading to the formation of the 32,514 km2 Nairobi Metropolitan Area. The region faces rapid urbanization challenges and lacks reliable data for urban planning. This study is aimed at monitoring the urban land cover using GIS and Remote Sensing techniques. Three different land cover maps produced at ten year epochs between 1995 and 2015 were used to evaluate and analyse urban growth visually and quantitatively. Spatial metrics indices and the Annual Urban Spatial Expansion Index were employed in the quantitative analyses. Urban land cover increased from 408.99 km2 to 763.79 km2 and to 2,187.82 km2 with annual growth rates of 8.4% and 17.2% in the two study period epochs. The areas had annual spatial expansion indices of 4.58% and 6.32% respectively in the periods 1995- 2000 and 2000-2015. Radial-axial expansion of built-up areas along transport routes led to the expansion of urban areas into agricultural and forest land. The region’s spatial plan forecasted an urban land cover of 1,717.72km2 by the year 2030, yet by the year 2015, the urban area was 2,187.82 km2. These growth characteristics will have an effect on the location of the proposed Cyber city, Knowledge-cum-Health city and the Aerotropolis, as proposed in the spatial planning concept of the NMR. The spatial analysis of the NMR growth trends could support in the urban growth management mechanisms and plans for sustainable development. Keywords GIS, Remote Sensing, Spatial urban growth, Nairobi Metropolitan Region, Sustainable development

and social development program, proposed flagship projects 1. Introduction to spur development across the country, one being the creation of six metropolitan areas covering the main cities. Urbanization is the process by which large numbers of The NMR project was the first be set up with a vision to grow people become permanently concentrated in relatively small and develop into a world class African metropolis capable of areas. It is induced by physical geography, living and creating sustainable wealth and offer a high quality of life to property costs, demand for more living space transportation, its residents, the people of Kenya and investors by the year and lack of proper planning policies [2]. The world is rapidly 2030 [7]. It was earmarked for rapid economic development urbanizing and cities are experiencing the dynamic processes since it plays an important role locally in the Kenyan of urbanization and globalization [8]. A large percentage of economy, regionally, as well as globally. the world’s population reside in metropolitan areas of Urban growth, being a land use/cover change phenomena developing countries [18] thereby raising a need of urban from a non-urban category to an urban category [3] is a planning. Kenya being one of Africa’s fastest urbanizing dynamic and complex phenomenon revealing economic countries had 215 urban centers in the year 2009 with development. Spatial temporal urban growth indicates the Nairobi as the capital city [10]. Twenty four (24) of these spatial and temporal dimensions of land cover/ use change at urban centers are located within the Nairobi Metropolitan the level of the urban landscape [5]. Urban land covers a Region (NMR). smaller area compared to other land covers but its impact to Previous land use/cover studies revealed that there was the surrounding environment is higher. It alters the land use unsustainable sprawl of Nairobi city [12]. It is estimated that pattern, land values and the intensity of site use [15]. Urban by the year 2030, 61.5% of the urban population will live growth combined with urban sprawl, inadequate within the NMR [13]. The Kenya Vision 2030, an economic infrastructure and land use planning have posed a challenge in the Nairobi Metropolitan Region-NMR [8]. * Corresponding author: [email protected] (Monica Machuma Katyambo) Urban growth generates a lot of problems and challenges Published online at http://journal.sapub.org/ajgis economically, socially and environmentally [13]. It has both Copyright © 2017 Scientific & Academic Publishing. All Rights Reserved negative and positive effects on the ecological and social

American Journal of Geographic Information System 2017, 6(2): 64-82 65

systems. Urbanization when ignored may intimidate caused by infill and by the vertical dimension of space was sustainable development. [6]. Urban expansion in the NMR not considered in assessing urban growth with regard to has induced significant socio-economic, environmental and settlement expansion. health challenges [4] raising a need of urban planning. This study answers questions such as where did urban Urban development is a large resource consumant that growth occur in the NMR between 1995 and 2005?; at what calls for proper planning and monitoring [10]. Lack of rate and how much land was converted to urban land cover information about urban spatial growth taking place in within the study period?; and which urban growth theory different parts of the metropolitan region has limited the explains the expansion. These questions were answered process of urban and regional planning and development with the aid of spatial metrics(total area-TA, class area-CA, management [8]. Urbanization in the NMR has led to high number of patches-NP, patch density-PD, largest patch-LP, land prices, low forest cover, and encroachment on largest patch index-LPI, mean patch size-MPS), urban conservation areas such as the Athi and Tana River growth indicators (annual urban spatial expansion catchments [9]. index-AUSEI) and through visual interpretation of the The Ministry of Nairobi Metropolitan Development resultant urban land cover maps. Spatial metrics are (MoNMeD) developed a spatial plan to determine how the numerical indices to describe structure and patterns of a NMR was to develop. The plan was to guide and coordinate landscape [14]. They quantify and describe the underlying development of infrastructural facilities and services and for structure and patterns of urban landscape from geospatial the specific control of the use and development of land [10]. data [16]. It proposed spatial modelling for the NMR through six new The study will enable spatial planners to determine the towns as new growth centres in the national economy as possibility to implement proposed development programs as well as to accommodate new activities meant to decongest it is possible to identify the available and lost land cover as Nairobi city [7]. The selection of the proposed new towns a result of new urban development. With this information, locations were based on accessibility from the road network, urban growth management mechanisms such as urban topography, availability of low productivity agricultural growth boundaries and delineation of land uses can be land, low interference with ecologically sensitive and implemented. conservation areas, and good potential for landscaping. This left out the crucial urbanization factor. Spatial planning requires information on the existing 2. Materials and Methods land-use/land-cover pattern, its spatial distribution and changes [17]. Urbanization determines the implementation To acquire the information needed in monitoring spatial of spatial planning proposals and of course has an impact on urban growth in the NMR, GIS and Remote Sensing sustainability of urban ecosystems. For the comprehensive technologies were used. land-use planning of urban areas, current and accurate land 2.1. The Study Area use information is vital for spatial planning and management [11]. Monitoring urban growth in the NMR is necessary for This study focuses on the Nairobi metropolitan region implementing appropriate strategies regarding the urban (NMR) comprising of four (4) out of the forty seven (47) planning decision making process and in redrawing urban counties in Kenya namely: Nairobi, , , policies in the NMR. and (Figure 1). The area is a highly urbanizing The main objective of this study was to monitor growth region with 24 urban centres. Nairobi city being the most in the urban land cover by analyzing the spatial-temporal dominant is the core of the metropolis. The region extends urban land use/cover changes that occurred in the region over an area of approximately 32514km2 that substantially before and after the creation of the Nairobi Metropolitan depend on Nairobi city for employment and social facilities. Area using GIS and remote sensing techniques. Landsat The NMR is strategically located as a central gateway to images were characterised, mapped and quantified to reveal the East and Central African region as well as its changes in the built-up land cover from which changes in positioning on the Northern corridor and the Cape to Cairo growth and pattern were evaluated and analysed. highway. International trunk roads A104, A109 pass The study focuses on the expansion of urban areas as sign through the area enabling movement of traffic to/from of economic development. Remote sensing and GIS to neighbouring landlocked countries of Uganda, technologies are used since they are a proper and effective Rwanda, South Sudan and Ethiopia. These characteristics tool to understand and present the phenomenon. Mapping present significant strengths for the region as it provides unplanned urban expansion provides a “picture” of where useful access points to various markets in Africa and this type of growth is occurring; its negative effects and particularly for the Indian Ocean islands and South Asia. suggests mitigation measures [17]. Accurate mapping of The Metro Vision 2030 divided the NMR into urban lands and monitoring urban expansions are significant foursub-regions: Core Nairobi (Nairobi county), the for optimum urban analysis in urban planning of Northern Metro (), the Eastern Metro metropolitan regions [1]. Spatial temporal urban growth () and Southern Metro ().

66 Monica Machuma Katyambo et al.: Spatial Monitoring of Urban Growth Using GIS and Remote Sensing: A Case Study of Nairobi Metropolitan Area, Kenya

Figure 1. Study Area, The Nairobi Metropolitan Area

2.2. Study Approach Figure 2 below graphically illustrates how the research was carried out to attain the main objective.

American Journal of Geographic Information System 2017, 6(2): 64-82 67

Figure 2. Study Approach 5 TM for 1995, Landsat 7 ETM+ for 2005 and Landsat 8 OLI for 2015 covering path 167 row 62, path 168 row 61,62 & 62, path 169 row 61 were used in land cover mapping (Figure 3 and Table 1).

2.4. Methods To capture the urban land use/cover in the study area from the imagery, data and image processing was done followed by temporal mapping and urban spatial change analysis. Digital image processing involved layer stacking, radiometric and geometric corrections, spatial enhancement, mosaicking, subsetting to curve out the AOI, and classification processes. The Scan Line Corrector (SLC) device in Landsat 7 malfunctioned in the year 2003 causing

stripping (gaps) in the imagery. To fill these gaps in the 2005 Figure 3. Landsat scenes overlay of the study area image, (destripping), focal analysis which is an iterative Table 1. Images used process, was adopted. The images were then enhanced to improve their visual appearance by applying a contrast SENSOR YEAR / PATH / ROW SOURCE adjustment technique, (Histogram Equalization). Landsat 5 1995, path 167 row 62, path 168 row USGS The images were classified into five feature classes TM 61,62& 62, path 169 row 61 namely: Built-up, Vegetation, Bare land, Water and Other Landsat 7 2005, path 167 row 62, path 168 row USGS lands using maximum likelihood classification algorithm in ETM+ 61,62 & 63, path 169 row 61 the supervised classification technique. The bare land class Landsat 8 2015, path 167 row 62, path 168 row USGS comprised the undeveloped or un-built-up areas in the study OLI 61,62& 62, path 169 row 61 area. The classification process had some challenges. Surfaces with similar spectral properties led to 2.3. Data misclassification of some land covers. Land covers that did Landsat imagery downloaded from United States not fall under built-up, vegetation, bare land or water classes Geological Survey (USGS) archives were used to reveal the were classified as other lands and the remaining cloud cover extent of urbanized land in the region. Five scenes of Landsat in the imagery after clearance was classified as water.

68 Monica Machuma Katyambo et al.: Spatial Monitoring of Urban Growth Using GIS and Remote Sensing: A Case Study of Nairobi Metropolitan Area, Kenya

Classification accuracy assessment was done by referencing urban growth indicator (the Annual Urban Spatial Expansion and assessing 40 randomly generated points of each class to Index), growth rates and seven spatial metrics. the Google earth high resolution images by synchronizing the ERDAS view with Google earth. The accuracy assessment report was then evaluated. 3. Results and Discussion The resultant classes were regrouped into built-up and non-built-up for easy analysis. The urban centre locations Post classification image differencing was used to show were singled out to capture built-up areas. Built-up areas the changes in urban land cover. The classified study area characterized urban areas and were curved out from the was mapped, quantified, analysed and evaluated. classified images. The twenty four urban centres within the 3.1. Classification NMR were singled out and the built up areas mapped to display the nature of changes in the years 1995, 2005 and The classified images had five land cover classes which 2015 classified images. Post classification comparison were presented in different tones (figure 4, 5 and 6). Surfaces approach was adopted for land use/cover change analysis. with similar spectral properties presented a challenge in the The analysis involved calculating the increase in the area of classification process leading to misclassification of some urbanized land over time and their growth rates using one land covers.

Figure 4. Classification Results 1995

American Journal of Geographic Information System 2017, 6(2): 64-82 69

Figure 5. Classification Results 2005

2 3.2. Accuracy Assessment to 3494.62km (10.9%) in 2015. Bare land and vegetation class areas reduced between 1995 and 2015 indicating a The generated accuracy assessment report showed the takeover. The areas reduced as e result of infill and extension Producer’s, User’s, Overall accuracies and Kappa developments. Bare lands were built-up while vegetation Coefficients (Table 2). was cleared to create space for development (Table 3). 2 Table 2. Accuracy Assessment results The built-up cover grew at a rate of 77.58km per year between 1995 and 2005. Between 2005 and 2015 the YEAR Overall Accuracy (%) Kappa Statistics built-up land cover grew at a rate of 146.46km2 per year. 1995 87.1 0.82 The built-up class area in Table 3 does not represent the 2005 87.62 0.81 urban area cover in the region since it excludes areas like 2015 87.94 0.85 open spaces, parks, airports, dump sites etc. and vegetation within the selected urban centres. On the other hand, the 3.3. Class Area Change Detection class reveals changes due to infill and extension From the classified images, land cover class areas in developments including the built up areas in the rural parts of square kilometres were generated (Table 3). Classification the NMR. The aim of this study was to monitor the urban results show that the built-up land cover increased from land cover by quantifying the changes that occurred between 1254.30km2 (3.9%) in 1995 to 2030.05km2 (6.3%) in 2005 1995 and 2015.

70 Monica Machuma Katyambo et al.: Spatial Monitoring of Urban Growth Using GIS and Remote Sensing: A Case Study of Nairobi Metropolitan Area, Kenya

Figure 6. Classification Results 2015

Table 3. NMR Land cover class areas for 1995, 2005 and 2015 Table 4. Regrouped land cover classes Land YEAR/ Area (km2) Category Descriptions Cover 1995 2005 2015 1 Built Up 1,254.30 2,030.05 3,494.62 Comprises all buildings in selected urban 2 Vegetation 14,722.22 14,465.15 13,783.75 centres (commercial & residential), roads (within the built up areas), and 3 Bare land 4,103.94 3,838.12 3,332.81 Built Up railway networks, impervious features, airports, 4 Water 5,421.15 5,398.17 5,350.74 vegetation, open spaces (Bareland), and dumpsites 5 Other Lands 6,657.71 6 ,427.87 6,197.31 within these centres, sport and leisure facilities, Totals 32,159.32 32,159.46 32,159.23 parks, etc. Comprises vegetation (forests, grasslands, 3.4. Description of Urban Land Use/Cover Changes agricultural) away from the urban centres, Water, Non- Built-Up roads (in rural areas and those connecting urban This study is all about urban expansion in the NMR only. centres) and Bare land not within the urban centres To define and quantify urban areas in the study area, land (open spaces) and the class other lands. cover classes were grouped into built-up and non-built –up (Table 4).

American Journal of Geographic Information System 2017, 6(2): 64-82 71

In a GIS environment, the 24 urban centres were curved showing built-up(urban) and non built-up(rural) land cover out from the classified images (using ArcGis 10.1). Maps for years 1995, 2005 and 2015 (Figures 7, 8 and 9).

Figure 7. 1995 Urban Area

72 Monica Machuma Katyambo et al.: Spatial Monitoring of Urban Growth Using GIS and Remote Sensing: A Case Study of Nairobi Metropolitan Area, Kenya

Figure 8. 2005 Urban Area

American Journal of Geographic Information System 2017, 6(2): 64-82 73

Figure 9. 2015 Urban Area

74 Monica Machuma Katyambo et al.: Spatial Monitoring of Urban Growth Using GIS and Remote Sensing: A Case Study of Nairobi Metropolitan Area, Kenya

Figure 10a. Nairobi County 1995 Urban land cover

Figure 10b. Nairobi County 2005 Urban land cover

American Journal of Geographic Information System 2017, 6(2): 64-82 75

Figure 10c. Nairobi County 2015 Urban land cover

Figure 11a. Kiambu and Machakos 1995 Urban Land Cover

76 Monica Machuma Katyambo et al.: Spatial Monitoring of Urban Growth Using GIS and Remote Sensing: A Case Study of Nairobi Metropolitan Area, Kenya

Figure 11b. Kiambu and Machakos 2005 Urban Land Cover

z Figure 11c. Kiambu and Machakos 2015 Urban Land Cover

American Journal of Geographic Information System 2017, 6(2): 64-82 77

Figure 12a. Kajiado 1995 Urban Land Cover

Figure 12b. Kajiado 2005 Urban Land Cover

78 Monica Machuma Katyambo et al.: Spatial Monitoring of Urban Growth Using GIS and Remote Sensing: A Case Study of Nairobi Metropolitan Area, Kenya

Figure 12c. Kajiado 2015 Urban Land Cover

Figure 13. NMR Built-Up Land Cover between 1995 and 2015 Figure 15. Built-Up and Non-Built-Up Land Cover

BUILT UP Urban mass 2.5 2.5 2 2 2187.82 2 2 1.5 1.5 1 BUILT UP 1 Urban mass 0.5 763.79

Thousands_km 0.5 0 Thousands_km 0 408.99 1995 2005 2015 1995 2005 2015

Figure 14. Graph showing the NMR built-up area between 1995 and 2015 Figure 16. Urban Spatial Expansion between 1995 and 2015

American Journal of Geographic Information System 2017, 6(2): 64-82 79

Table 5. NMR urban area cover and changes

YEAR/URBAN AREA(km2) CHANGE (km2) COUNTY AREA 1995 2005 2015 1995-2005 2005-2015 Nairobi 700.50 302.15 492.62 700.50 172.47 207.88 Kiambu 3,275 29.82 98.30 951.40 68.48 853.10 Machakos 6,281 21.72 97.23 379.17 75.51 281.94 Kajiado 2,1903 37.30 75.64 156.75 38.34 81.11 TOTAL 32,159.50 408.99 763.79 2,187.82 354.80 1,424.03

Table 6. NMR Urban Land Cover in square kilometres

Urban Centre YEAR/AREA (km2) Spatial Change (km2) County 1995 2005 2015 1995-2005 2005-2015 Nairobi Nairobi 320.15 492.62 700.50 172.47 207.88 Kiambu 1.47 3.55 2.75 1.59 2.71 65.51 1.12 59.25 Lower Kabete ------ 0.62 6.02 5.4 Kikuyu ------8.77 60.72 8.77 45.93 Kiambu Thogoto ------ 2.62 19.98 16.58 2.79 16.79 805.51 691.58 13.05 37.47 46.48 Githunguri 0.89 2.41 17.66 1.52 15.25 Gatundu 0.42 0.60 2.00 0.18 1.40 Total 29.82 98.30 951.40 68.48 853.10 Machakos 2.47 10.67 44.97 8.20 34.30 Mavoko 14.83 76.42 61.59 Kathiani 1.05 3.52 325.40 2.47 245.45 Machakos Mulolongo, Syokimau ------Tala 3.06 5.96 7.25 2.98 0.30 0.31 0.58 1.54 0.27 0.96 Total 21.72 97.23 379.17 75.51 281.01 Kajiado 1.40 5.49 17.16 4.09 11.67 Iibissil 0.18 0.33 1.17 0.15 0.84 Isinya 1.49 7.52 13.71 6.03 6.19 2.82 3.14 10.54 2.04 7.40 0.16 2.20 9.85 Kajiado 9.46 19.31 2.27 99.21 44.94 14.07 16.34 10.59 Ngong 5.83 16.42 0.41 Oloitoktok 0.16 0.81 9.99 0.65 9.42 Sultan Hamud 1.73 4.08 4.97 2.35 0.89 Total 37.30 75.64 156.75 38.43 81.11 NMR urban Total 408.99 763.79 2,187.82 354.80 1,424.03

Table 7. Annual Urban Spatial Expansion Index for periods 1995 - 2005 and 2005- 2015

Year Urban area (km2) Spatial expansion (km2) AUSEI (%)/UEII 1995 408.99 ------2005 763.79 354.80 4.65 2015 2,187.82 1424.03 6.51

80 Monica Machuma Katyambo et al.: Spatial Monitoring of Urban Growth Using GIS and Remote Sensing: A Case Study of Nairobi Metropolitan Area, Kenya

Table 8. Description of the NMR using spatial metrics

Year/ TA CA LP MPS LULC NP PD LPI Metrics (km2) (km2) (km2) (km2) Built-up 1995 32159.50 408.99 24 0.05 320.15 78.3 17.78 Built-up 2005 32159.50 763.79 24 0.03 492.62 64.5 31.82 Built-up 2015 32159.50 2,187.82 17 0.008 700.5 32.0 136.74

Table 9. Analysis of built-up area expansion based on CA metrics

Growth Annual rate of Study Period CA Change (km2) Change % Average Rate (%) change (km2) 1995-2005 354.80 86.80 10 8.70 14 36.85 2005-2015 1,424.03 186.40 10 18.60 142.40

The NMR sub-regions were curved out to enable the reduced fragmentation of the built-up area in the region. A analysis of urban growth at sub-region level (Figures 10, 11, decrease in PD from 0.05 in 1995 to 0.008 in 2015 indicated and 12). the expansion of patches as a result of rapid continuous development and a decrease in the number of patches (Table 3.5. NMR Land Use/Cover Status for 1995-2015 9). The urban area captured over three different years was Nairobi city is the largest patch (LP) in the NMR. The quantified and compared (Figure 7, 8, 9, Table 5 and 6). Of significant decrease in the largest patch index (LPI) in the the 32159.5 km2 NMR area, urban centres covered 408.99 NMR indicated the expansion of the other urban centres or km2 (1.27%), 763.79 km2 (2.38%), and 2187.82 km2 (6.80%) patches and a reduction in the non-built-up area in the region in 1995, 2005 and 2015 respectively (Figure 13 and 15). (Figure 17). Rural areas covered 31750.51km2 (98.73 %), 31395.71 Continuous urban growth and development in the built-up km2 (97.62%) and 29971.68km2 (93.20 %) in 1995, 2005 and area led to an increase in MPS and a reduction in 2015 respectively (Figure 15). fragmentation hence a reduction in the NP and PD caused by merging of the patches into larger patches. These patches 3.6. The NMR Urban Growth Analysis grew as a result of their proximity to the dominant patch. NP The Concentric zone, axial development and multiple reduced from 24 to 17 and PD from 0.05 to 0.008 (Table 8). nuclei theories explain urban growth in the NMR. The rates of urban land cover expansion varied during the Infrastructure projects like roads, electricity and water 1995-2005 and 2005-2015 time periods (Table 9). supply led to the expansion of the urban centres resulting According to the CA metric, the total built-up area from improved accessibility, service delivery and living increased by 1778.83km2 between 1995 and 2015.With standards in general. Built-up expansion occurred in a regard to 1995, built-up land cover expanded by encroaching radial-axial manner along transport routes away from the on non built-up land cover by 1.12% and 4.49% in 2005 and core Nairobi city towards the urban centres and away from 2015 respectively (Table 10). 1778.83 km2non urban land the urban centres towards Nairobi (Figures 7, 8 and 9). cover was converted to urban land cover within the study period. Table 10. Built-up encroachment on non-built up land with regard to 1995

Year 2005 2015 % encroachment on non-built up land 1.12% 4.49%

The urban land mass expanded rapidly between 2005 and 2015 (1424.03km2) compared to between 1995 and 2005 (354.80km2) (Figure 16). Urban growth was analysed using the indicator system. AUSEI was used to determine temporal changes in urban areas and growth rates (Table 8). Spatial metrics were used to describe and quantify changes in urban land cover (9). The number of patches (NP) metric quantifies the number of individual urban centres. There were 24 patches in 1995 and in 2005. NP reduced to 17 as a result of urban centres expanding towards each other and merging into continuous Figure 17. Graph showing NMR largest patch index (LPI) urban land cover. Changes in LPI and PD (fragmentation metrics) imply a decrease in spatial heterogeneity. There is The expansion of Kangundo and Tala urban centres

American Journal of Geographic Information System 2017, 6(2): 64-82 81

demonstrate the effect of topography on urban growth. areas. The expansion happened along transport routes from Kangundo town has not expanded extensively as compared Nairobi city to other urban centres and from the urban to Tala trading centre due to the nature of the terrain at its centres to Nairobi city. The expansion was determined by location. By the year 2015, Nairobi County was 100% environmental, social-economic, political, and technological urbanised with an urban land cover of 700.50km2. It was factors. The government economic policy, improvements in followed by the Northern metro (Kiambu county) and then infrastructure, part VI of the Physical Planning Act(CAP 286) the Eastern metro (Machakos county) with urban areas of of 1996, high values of land at the CBD’s i.e at the core 951.40km2 and 379.17km2 respectively. The Southern metro Nairobi city and at the NMR urban centres, topography, (Kajiado county) is the least urbanised region in the NMR population increase in the region and people opting to live on with 156.75km2 of urban land cover and a patch density of less congested affordable land encouraged spatial urban 4.47 (Table 11). growth. Urban growth in the region does not affect the location of the NMR spatial plan proposed Transport city and Amboseli new town. The Cyber city, Knowledge cum 4. Conclusions Health city and the Aerotropolis need to be relocated due to Significant urban growth influenced by both spatial and urbanization. The spatial plan forecasted an urban land cover 2 non-spatial factors occurred in the NMR within the study of 1,717.72km by the year 2030, yet by the year 2015, the 2 period especially after formation of the Ministry of Nairobi NMR urban area was 2,187.82 km . Lack of information Metropolitan Development (MoNMeD) in the year 2008. about spatial urban growth in NMR like up to date maps Urban expansion in the NMR was dominated by extension prepared from high resolution satellite imagery denies developments and sprawl in a radial-axial manner at the planners a synoptic view of built up and non-built up areas in Nairobi city periphery onto agricultural land in the rural the region needed to formulate sustainable urban development strategies.

Table 11. Description of the NMR regions using spatial metrics

Metro Region Spatial Metric Nairobi Eastern Northern Southern Total (Core) Metro (Machakos) Metro (Kiambu) Metro (Kajiado) Region Area (km2) 700.50 6,281 3,275 2,1903 32,159.50 CA (urban area (km2) 1995 320.15 21.72 29.82 37.30 408.99 2005 492.62 97.23 98.30 75.64 763.79 2015 700.50 379.17 951.40 156.75 2,187.82 NP (n) 1995 1 5 8 10 24 2005 1 5 8 10 24 2015 1 4 5 7 17 PD (%) 1995 0.31 23.02 12.23 26.81 2005 0.2 5.14 5.74 13.22 2015 0.14 1.05 0.52 4.47 MPS (km2) 1995 320.15 4.34 8.18 3.73 2005 492.62 19.45 17.42 7.56 2015 700.50 94.79 190.56 22.39 LP (km2) 1995 320.15 14.83 13.05 14.07 2005 492.62 76.42 37.47 16.34 2015 700.50 325.40 805.51 99.21 LPI (%) 1995 100 68.3 43.8 37.7 2005 100 78.6 38.1 25.5 2015 100 85.8 84.6 63.3

82 Monica Machuma Katyambo et al.: Spatial Monitoring of Urban Growth Using GIS and Remote Sensing: A Case Study of Nairobi Metropolitan Area, Kenya

Urban growth monitoring in the NMR will guide resource [7] MoNMeD. Final Spatial Planning Concept for Nairobi managers in making prudent decisions and for appropriate Metopolitan Region. Nairobi: GoK, 2012. allocation of resources. The maps will guide the NMR [8] MoNMeD. Nairobi Metro 2030. Nairobi: Government of planners in defining urban growth boundaries and delineate Kenya, 2008. land uses. [9] MoNMeD. Request for Proposals (RFP) to Develop a Spatial Planning Concept for the Nairobi Metropolitan region, Nairobi. Nairobi: GoK, 2009. ACKNOWLEDGEMENTS [10] MoNMeD. The first edition of the Spatial Planning Concept. We acknowledge the Regional Centre for Mapping Nairobi: GoK, 2012. Resources and Development (RCMRD) for their support [11] Mubea, Kenneth. W. "Monitoring Land-Use Change in with the satellite imagery used in this study. (Kenya) Using Multi-Sensor Satellite Data." Advances in Remote Sensing., 2012: 1,74-84.

[12] Mundia, Charles. Ndegwa, and Aniya. M. "Analysis of Land Use Land Cover Changes and Urban Expansion of Nairobi City Using Remoote Sensing." International Journal of REFERENCES Remote Sensing, 2005: 2831-2849. [1] Al-sharif et al. "Quantitative analysis of urban sprawl in [13] Omwenga, Mairura. "Nairobi - Emerging Metropolitan Tripoli using Pearson’s Chi-Square statistics and urban Region: Development Planning and Management expansion intensity index." IPOScience (IOP Publishing), Opportunities and Challenges." Nairobi, 2010. 2006. [14] O'neil, R. V., Krummel, J. R., Gardner, R.H., Sugihara, G., [2] Batta, B. Analysis of Urban Growth and Sprawl from Remote Jackson, B,. "Indices of landscape pattern." Landscape Sensing Data. Springer-Verlag Berlin Heidelberg, 2010. ecology 1(3), 1988: 53-162. [3] Batty, Michael. "Geopatial analysis and modelling of urban [15] Paul. N, Balchin, Bull Gregory. H, and Hieve. Jeffrey. L. structure and dynamics." The Geojournal, 1999. Urban Land Economics, Fifth Edition. Palgrave Macmillan, [4] Bousquet, Anne. Urban development in Kenya: Towards 1995. inclusive cities. No. 2008. Nairobi: French Institute for [16] Pham, H.M., and Y. Yamaguch. "Urban growth and change Research in Africa, 2008. analysis using remote sensing and spatial metrics fom 1975 to [5] Cheng, Jianquan. Modelling Spatial and Temporal Urban 2003 for Hanoi, Vietnam." International journal of remote Growth. Enschede: Febodruk BV, Enschede, The sensing, 32(7), 2011: 1901-1915. Netherlands, 2003. [17] Ravindra, K.V. "Application of remote sensing and GIS [6] Dubovyk. O., Sliuzas, R, Flacke. J. "Spatial temporal technique for efficient urban planning in India." 2008. modelling of informal settlement development in Sancaktepe [18] UN HABITAT. Save Our Cites. EARTH ACTION, 1996. district, Istanbul, Turkey." Isprs journal of photogrammetry and remote sensing (66)2, 2011: 235-246.